Desenvolvimento de uma metodologia moderna de projeto de veículos

• Ano de defesa: 2007
• Autor(a) principal: Leal, Marcus de Freitas
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Engenharia Mecânica; Engenharias; UFU
• Palavras-chave em Português: Dinâmica de veículos; Modelagem computacional; Validação experimental; Otimização numérica; Método de superfícies de resposta; Veículos a motor - Dinâmica; Vehicle dynamics; Computational modeling; Experimental validation; Numerical optimization; Meta-modeling; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/14721; https://doi.org/10.14393/ufu.te.2011.1

The study of vehicle dynamics has been widely improved by the use of numerical simulation, in particular related to the several possibilities associated to complicated and representative models. In the context of the evolution of optimization methods applied to vehicle dynamics, a methodology development able to establish the design actions and requirements related to the vehicle has been proposed. Several numerical tools were used in the system modeling and analysis through the multibody technique, finite element method and numerical optimization. An intense effort was made to build a computational model of a real vehicle which was similar to the final intended design. That involves, three-dimensional component models, validation of the dynamic behavior using experimental tests under several operation conditions, experimental determination of physical properties like inertia tensor, stiffness and damping. The vehicle model was refined by considering the chassis flexibility. After the modeling and validation steps, the main aspects of the vehicle dynamics were analyzed taking into account comfort and performance. These aspects were used in the formulation of multidisciplinary and multi-objective optimization problems in order to obtain a better vehicle configuration. The design variables used in the problems are related to characteristic properties of components and systems, in which were applied meta-modeling techniques. At last, we have the final considerations and conclusions of the proposed and accomplished work.